The present invention relates to Coriolis-type mass flowmeters. More particularly, the invention relates to such a flowmeter which comprises at least one flowtube and a number of strain-sensing transducers for measuring the driven and Coriolis deflections of the flowtube.
Coriolis mass flowmeters typically comprise one or more geometrically precise flowtubes which are made of a strong yet resilient material such as stainless steel or titanium. The flowtube is mounted in a supporting structure having end connections, such as a casting or a pipe, to protect the flowtube and to provide for the attachment of the flowtube to external process piping. Coriolis flowmeters usually also comprise a number of force drivers for vibrating the flowtube in one of its modes of vibration, such as its first bending mode of vibration, and a plurality of motion sensors for measuring the vibratory deflections of the flowtube. The mass of the fluid passing through the vibrating flowtube generates Coriolis forces that act on the flowtube and cause it to deflect into a unique and characteristic shape, which is referred to herein as the xe2x80x9cCoriolis deflectionxe2x80x9d.
In many prior art Coriolis flowmeters, the motion sensors typically comprise two magnet/coil pairs: one located near the upstream end of the flowtube and the other located near the downstream end of the flowtube. Thus instrumented, the differences between the signals produced by the upstream and the downstream motion sensors as a result of the Coriolis deflection of the flowtube can be used to determine the mass flow rate of the fluid passing through the flowmeter, as is well known to those skilled in the art.
However, the use of magnet/coil pairs for the motion sensors is problematic for several reasons. First, the magnets and coils represent a significant mass loading on the flowtube which tends to alter the frequency response of the flowtube to the Coriolis forces. This is especially true if the flowmeter comprises two motion sensors which are displaced from each other along the length of the flowtube. Second, the Coriolis-induced deflection of the flowtube typically increases toward the ends of the flowtube, which is where the detrimental boundary condition effects typically occur. This creates a conflict between positioning the sensors close enough to the ends of the flowtube where the Coriolis induced deflections are large, yet far enough away from the ends to minimize any errors due to boundary condition effects. Third, individual magnet/coil pairs require significant care during assembly to achieve the necessary precision in positioning along the length of the flowtube, mass matching for accurate balance, and velocity output signal matching.
In accordance with the present invention, these and other disadvantages in the prior art are overcome by providing a Coriolis mass flowmeter which comprises at least one flowtube through which a fluid to be measured is allowed to flow and at least one force driver for vibrating the flowtube in at least one mode of vibration of the flowtube, such as its first bending mode of vibration. This vibration causes a driven deflection of the flowtube, and the fluid flowing through the vibrating flowtube generates Coriolis forces which cause a Coriolis deflection of the flowtube. Accordingly, the invention further comprises a plurality of strain-sensing transducers connected to the flowtube for generating signals representative of the driven deflection of the flowtube and the Coriolis deflection of the flowtube, and a signal processing circuit connected to the strain-sensing transducers for producing a measure of the flow rate of the fluid from the signals generated by the strain-sensing transducers.
Furthermore, it has been discovered that both the driven deflection and the Coriolis deflection of the flowtube may be sensed at a single location along the length of the flowtube, which is preferably near the mid-span centerline of the flowtube. Also, when subject to the driven deflection, for example when driven in either its first bending mode or first radial mode of vibration, the flowtube experiences a maximum displacement and consequently a local maximum normal strain, but little or no shear strain, near its mid-span centerline. Moreover, when subject to the Coriolis deflection, the flowtube experiences a local maximum shear strain, but little or no normal strain, near its mid-span centerline.
Therefore, the plurality of strain-sensing transducers preferably includes at least one normal strain-sensing transducer for detecting the normal strain on the flowtube resulting from the driven deflection of the flowtube, and at least one shear strain-sensing transducer for detecting the shear strain on the flowtube resulting from the Coriolis deflection of the flowtube. Moreover, both the normal strain-sensing transducer and the shear strain-sensing transducer are preferably mounted on the flowtube near its mid-span centerline. In addition, the normal and shear strain-sensing transducers optimally each comprise a simple, commercially available strain gage element. Furthermore, the strain gage elements of the normal and shear strain-sensing transducers are preferably mounted on a single substrate which is affixed to the flowtube to thereby simply the manufacture of the flowmeter.
Thus, it can be seen that the reliability and accuracy of the flowmeter of the present invention are improved by the use of motion sensors which comprise simple strain-sensing transducers. Since the present invention enables the use of only one set of sensors, rather than two magnet/coil pairs disposed along the length of the flowtube, the number of critical components in the flowmeter is accordingly reduced. This results in fewer components that can fail and also reduces the manufacturing cost of the flowmeter. Also, positioning the strain-sensing transducers near the mid-span centerline of the flowtube improves the accuracy of the flowmeter for several reasons. First, the reduced weight of the transducers minimizes the affect the sensors may have on the frequency response of the flowtube. Second, the mid-span sensing location is least affected by inaccuracies caused by boundary condition effects at the ends of the flowtube. Third, the strain gage-type sensors preferred for the present invention are highly linear and have high frequency response compared to magnet/coil pairs. Therefore, the Coriolis mass flowmeter of the present invention is generally more reliable than and capable of achieving much improved accuracy over prior art flowmeters.
These and other objects and advantages of the present invention will be made apparent from the following detailed description, with reference to the accompanying drawings. In the drawings, the same reference numbers are used to denote similar components in the various embodiments.